!python --version

Python 3.11.7

# Install libraries
# !pip install scikit-posthocs

# Import libraries

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
import scipy.stats as stats
import plotly.express as px
import ipywidgets as widgets
from IPython.display import display
import scikit_posthocs as sp
import plotly.io as pio
pio.renderers.default='notebook'

%matplotlib inline

# Display all columns in the dataframe
pd.set_option('display.max_columns', None)

# Set the display format for floats
pd.set_option('display.float_format', lambda x: '%.3f' % x)

# Set the figure size
plt.rcParams["figure.figsize"] = (10, 6)

# Set style for plots
sns.set_style("whitegrid")

# Ignore all warnings
import warnings
warnings.filterwarnings("ignore")
warnings.warn("this will not show")

pd.__version__

'2.1.4'

# Load csv
df = pd.read_csv('tmdb-movies.csv')

df.head()

df.shape

(10866, 21)

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10866 entries, 0 to 10865
Data columns (total 21 columns):
 #   Column                Non-Null Count  Dtype  
---  ------                --------------  -----  
 0   id                    10866 non-null  int64  
 1   imdb_id               10856 non-null  object 
 2   popularity            10866 non-null  float64
 3   budget                10866 non-null  int64  
 4   revenue               10866 non-null  int64  
 5   original_title        10866 non-null  object 
 6   cast                  10790 non-null  object 
 7   homepage              2936 non-null   object 
 8   director              10822 non-null  object 
 9   tagline               8042 non-null   object 
 10  keywords              9373 non-null   object 
 11  overview              10862 non-null  object 
 12  runtime               10866 non-null  int64  
 13  genres                10843 non-null  object 
 14  production_companies  9836 non-null   object 
 15  release_date          10866 non-null  object 
 16  vote_count            10866 non-null  int64  
 17  vote_average          10866 non-null  float64
 18  release_year          10866 non-null  int64  
 19  budget_adj            10866 non-null  float64
 20  revenue_adj           10866 non-null  float64
dtypes: float64(4), int64(6), object(11)
memory usage: 1.7+ MB

# First look - Descriptive statistics
df.describe().T

def replace0(df:pd.DataFrame,feature:str):
    """Replace 0s with NaNs in a dataframe column and returns the zero counts."""
    before = df.query(f'{feature} == 0').shape[0]
    df[feature].replace({0: np.nan}, inplace=True)
    after = df.query(f'{feature} == 0').shape[0]
    return f"the number of zeros in {feature} column : {before} -> {after}"

# Display a sample
df.sample(1)

# Drop columns
drop_column_list = ['imdb_id','budget','revenue','homepage','tagline','overview','keywords']
display(df.shape)
df.drop(columns=drop_column_list,inplace=True)
display(df.shape)

(10866, 21)

(10866, 14)

# Display a sample
df.sample(1)

# Check if the Dataset have any Duplicate

df.duplicated().value_counts()

False    10865
True         1
Name: count, dtype: int64

# Drop Duplicates

df.drop_duplicates(inplace=True)

# Display shape

df.shape

(10865, 14)

# Display descriptive statistics
df.describe().T

# replace 0 with np.nan in the runtime column
replace0(df,'runtime')

'the number of zeros in runtime column : 31 -> 0'

# replace 0 with np.nan in the budget_adj
replace0(df,'budget_adj')

'the number of zeros in budget_adj column : 5696 -> 0'

# replace 0 with np.nan in the revenue_adj
replace0(df,'revenue_adj')

'the number of zeros in revenue_adj column : 6016 -> 0'

# Check the sum of Missing Values per column

df.isnull().sum()

id                         0
popularity                 0
original_title             0
cast                      76
director                  44
runtime                   31
genres                    23
production_companies    1030
release_date               0
vote_count                 0
vote_average               0
release_year               0
budget_adj              5696
revenue_adj             6016
dtype: int64

# Check the Percentage of Missing Values

df.isnull().sum() / df.shape[0] * 100

id                      0.000
popularity              0.000
original_title          0.000
cast                    0.699
director                0.405
runtime                 0.285
genres                  0.212
production_companies    9.480
release_date            0.000
vote_count              0.000
vote_average            0.000
release_year            0.000
budget_adj             52.425
revenue_adj            55.370
dtype: float64

# Plot the Percentage of Missing Values

ax = (df.isnull().sum() / df.shape[0] * 100).plot(kind='barh')
plt.bar_label(ax.containers[0], fmt='%.2f%%');

# Display a sample
df.release_date.sample(1)

31    3/4/15
Name: release_date, dtype: object

# Convert str to date format
df['release_date'] = pd.to_datetime(df['release_date'])

# Control a sample
df.release_date.loc[10561]

Timestamp('1986-09-23 00:00:00')

# List of columns which data types are object
df.select_dtypes(include=["object"]).columns

Index(['original_title', 'cast', 'director', 'genres', 'production_companies'], dtype='object')

# Descriptive statistics of object columns
df.select_dtypes(include=["object"]).describe().T

# make a copy
df0 = df.copy()

def explode(df_:pd.DataFrame,feature:str):
    """Split and expand a column of delimited values into multiple rows."""
    df_temp = df_.copy()
    df_temp[feature] = df_temp[feature].str.split('|')
    return df_temp.explode(feature)

# Get one sample
df.sample(1)

# Explode dataframe to have each genre as a seperate row

df_cast_explode = explode(df,'cast')
df_cast_explode.head(3)

# the number of null values

df_cast_explode.cast.isna().sum()

76

# Drop null values

display(df_cast_explode.shape)
df_cast_explode.dropna(subset=['cast'], inplace = True)
display(df_cast_explode.shape)

(52644, 14)

(52568, 14)

# the number of duplicated records

df_cast_explode.duplicated().sum()

19

# Drop duplicates

df_cast_explode.drop_duplicates(inplace=True)
display(df_cast_explode.shape)

(52549, 14)

# the unique number of cast members in the dataset

df_cast_explode.cast.nunique()

19026

# Top 5 cast members

cast_counts = df_cast_explode['cast'].value_counts()[:5]
cast_counts

cast
Robert De Niro       72
Samuel L. Jackson    71
Bruce Willis         62
Nicolas Cage         61
Michael Caine        53
Name: count, dtype: int64

# Visualization of top 5 members

ax = sns.barplot(x=cast_counts.index, y=cast_counts.values, palette = 'Blues_r')
plt.xlabel('Cast')
plt.ylabel('Count')
plt.title('Top 5 cast members by frequency')
plt.xticks(rotation=45)
for container in ax.containers:
    plt.bar_label(container);

# Explode dataframe to have each genre as a seperate row
df_genres_explode = explode(df,'genres')
df_genres_explode.head(3)

# Look at its shape
df_genres_explode.shape

(26978, 14)

# Genres in the dataset
df_genres_explode.genres.unique()

array(['Action', 'Adventure', 'Science Fiction', 'Thriller', 'Fantasy',
       'Crime', 'Western', 'Drama', 'Family', 'Animation', 'Comedy',
       'Mystery', 'Romance', 'War', 'History', 'Music', 'Horror',
       'Documentary', 'TV Movie', nan, 'Foreign'], dtype=object)

# the number of null values in genres
df_genres_explode.genres.isna().sum()

23

# Drop nulls if exist in genres column
display(df_genres_explode.shape)
df_genres_explode.dropna(subset=['genres'], inplace = True)
display(df_genres_explode.shape)

(26978, 14)

(26955, 14)

# Genres count in the dataset
df_genres_explode.groupby('genres').count()[['id']]

# Genres count in the dataset sorted descending
df_genres_explode['genres'].value_counts()

genres
Drama              4760
Comedy             3793
Thriller           2907
Action             2384
Romance            1712
Horror             1637
Adventure          1471
Crime              1354
Family             1231
Science Fiction    1229
Fantasy             916
Mystery             810
Animation           699
Documentary         520
Music               408
History             334
War                 270
Foreign             188
TV Movie            167
Western             165
Name: count, dtype: int64

# the visualization of genres count in the dataset sorted descending
ax = sns.countplot(x='genres',data=df_genres_explode, order=df_genres_explode['genres'].value_counts().index, palette = 'Blues_r')
plt.xticks(rotation=45)
for container in ax.containers:
    plt.bar_label(container);

# Genres popularity mean by year
df_genres_explode.groupby(['release_year','genres'])[['popularity']].mean().reset_index().head(20)

# Line plot of genres popularity mean by year 
df2 = df_genres_explode.groupby(['release_year','genres'])[['popularity']].mean().reset_index()
ax = sns.lineplot(x='release_year',y='popularity',hue='genres',data=df2)
sns.move_legend(ax,'upper left', bbox_to_anchor=(1,1))

g = sns.FacetGrid(df_genres_explode, col="genres", col_wrap=5, height=3, aspect=1.3)
g.map(sns.lineplot, "release_year", "popularity")
g.set_titles("{col_name}")
g.tight_layout()

<seaborn.axisgrid.FacetGrid at 0x23f2ef8e7d0>

# Create an interactive line plot with plotly express
fig = px.line(df2, x='release_year', y='popularity', color='genres', width= 1000, height = 600)

# Show the interactive plot
fig.show()

# Select the most popular genre of the years
idx = df2.groupby('release_year')['popularity'].idxmax()

# Select those rows from the original dataframe
most_popular = df2.loc[idx]

# What are the most popular genres each year?
most_popular

# Get the number of movies by grouping according to release year.
df_year = df.groupby('release_year').size()
df_year.head(3)

release_year
1960    32
1961    31
1962    32
dtype: int64

# Line plot of number of movies across years.
sns.lineplot(x=df_year.index,y=df_year.values, label= 'Number of movies', color='green')
plt.xlabel("Release year")
plt.ylabel("Number of movies")
plt.title("Number of movies by release year")
plt.show()

# https://plotly.com/python/line-charts/

fig = px.line(df_year, x=df_year.index, y=df_year.values, labels={'y':'Number of movies'}, title='Number of movies by release year', width= 1000, height=600)
fig.update_layout(hovermode="x")
fig.show()

# The histogram of popularity
sns.histplot(x='popularity',kde=True, data=df, bins=20);

# Descriptive statistics of popularity
df.describe()[['popularity']].T

# The boxplot of popularity
sns.boxplot(x='popularity', data=df)

<Axes: xlabel='popularity'>

# Histograms of popularity by genre
plt.figure(figsize=(20,15))
for count,genre in enumerate(df_genres_explode.genres.unique()):
    plt.subplot(5,5,count + 1)
    sns.histplot(x='popularity',kde=True, data=df_genres_explode.query('genres == @genre'), bins=20)
    plt.title(genre)
plt.tight_layout()

# Genre popularity skewness
for count,genre in enumerate(df_genres_explode.genres.unique()):
    print(f"{genre}: {df_genres_explode.query('genres == @genre').popularity.skew()}")

Action: 8.519559576213021
Adventure: 7.544246925782778
Science Fiction: 8.346574445070894
Thriller: 12.31808653463584
Fantasy: 3.3442974777303105
Crime: 4.141872354110435
Western: 6.173147812882716
Drama: 9.26647507842431
Family: 3.399000948901938
Animation: 3.2776512678707967
Comedy: 3.85839662551183
Mystery: 4.033738855312595
Romance: 3.0004699872067215
War: 4.07009677907743
History: 5.559047399136068
Music: 3.404349210716752
Horror: 5.209936328110234
Documentary: 1.7119359791142117
TV Movie: 2.8382606971691686
Foreign: 1.1906528356317898

# Looking at the mean and the median side by side
df_genres_explode.groupby('genres')[['popularity']].agg(['mean','median'])

# Median popularity of genres
df_median_popularity = df_genres_explode.groupby('genres')['popularity'].agg(['median'])
df_median_popularity

# Make genre groups to be calculated in the statistical test
genre_groups = [group['popularity'].values for name,group in df_genres_explode.groupby('genres')]

# Apply Kruskal Wallis test to genre groups
statistic, p_value = stats.kruskal(*genre_groups)

# display Kruskal Wallis test results
print(f'Kruskal-Wallis Statistic: {statistic}')
print(f'p-value: {p_value}')

Kruskal-Wallis Statistic: 1586.9103452441982
p-value: 0.0

# Perform hyptoheses test
if p_value < 0.05:
    print('There is a statistically significant difference in median popularity across genres')
else:
    print('There is no statistically significant difference in median popularity across genres')

There is a statistically significant difference in median popularity across genres

# https://www.statology.org/dunns-test-python/
# To find out which groups have different medians after a Kruskal-Wallis test, we can use Dunn’s Test for pairwise comparisons.
posthoc_result = sp.posthoc_dunn(df_genres_explode,val_col='popularity',group_col='genres',p_adjust='bonferroni')

# dataframe that shows True if the difference is statistically significant, shows False if there is not statistically significant difference between medians of the pairs.
posthoc_result_p = posthoc_result < 0.05

# Heatmap of genre pairs showing whether there is a statistically significant difference between pair medians.

plt.figure(figsize=(20, 20)) 
sns.heatmap(posthoc_result_p, annot=True, cmap='Blues', fmt="", cbar=False) 
plt.xlabel("Genres")
plt.ylabel("Genres")
plt.title("Post-hoc Dunn's test results");
# True means there is statistically significant difference between median popularity of the two genres.
# False means  there is no statistically significant difference between median popularity of the two genres.

# Creating a new df to be used in plotly interactive hover template
# Create an empty numpy array with the same shape as df_median_popularity
new_array = np.empty((len(df_median_popularity), len(df_median_popularity)))
new_array = new_array.astype(str)

# Fill the new array with the value of the pairs
for i, x in enumerate(df_median_popularity.index):
    for j, y in enumerate(df_median_popularity.index):
        new_array[i][j] = f'({x}:{df_median_popularity.loc[x][0]} {y}:{df_median_popularity.loc[y][0]})'

# Convert the new array to a pandas dataframe
new_df = pd.DataFrame(new_array, index=df_median_popularity.index, columns=df_median_popularity.index)

# Display a sample
new_df.sample(1).iloc[0,0]

'(Documentary:0.1422399999999995 '

# Interactive heatmap showing median pairwise comparisons.

fig = px.imshow(posthoc_result < 0.05, color_continuous_scale='Blues')
fig.update_xaxes(side="top")
fig.update_layout(title="Post-hoc Dunn's test results", title_x=0.5, width=1000, height=1000)
fig.update_coloraxes(showscale=False)
fig.update_traces(text=new_df, texttemplate="%{text}")
fig.update_traces(hovertemplate='%{x} - %{y} <br> %{text}')
fig.show()
# Blue color means there is statistically significant difference between median popularity of the two genres.
# White color means  there is no statistically significant difference between median popularity of the two genres.

# First look at the runtime-popularity scatterplot across all movies

plt.figure(figsize=(8, 6))
sns.scatterplot(data=df, x="runtime", y="popularity")
plt.title(f"All Movies")
plt.xlabel("Runtime")
plt.ylabel("Popularity")
plt.show();

# Display the runtime-popularity scatterplot - genres breakdown

plt.figure(figsize=(8, 6))
g = sns.FacetGrid(df_genres_explode, col="genres", col_wrap=5, height=3, aspect=1.3)
g.map(sns.scatterplot, "runtime", "popularity")
g.set_titles("{col_name}")
g.tight_layout();

<Figure size 800x600 with 0 Axes>

# Make it interactive. Create a dropdown widget for genre selection

genre_dropdown = widgets.Dropdown(
    options=df_genres_explode["genres"].unique(),
    description="Select Genre:",
    disabled=False,
)

def plot_scatterplot(genre):
    plt.figure(figsize=(8, 6))
    sns.scatterplot(data=df_genres_explode[df_genres_explode["genres"] == genre],
                    x="runtime", y="popularity")
    plt.title(f"{genre} Movies")
    plt.xlabel("Runtime")
    plt.ylabel("Popularity")
    plt.xlim(0, df_genres_explode["runtime"].max())
    plt.ylim(0, df_genres_explode["popularity"].max())
    plt.show()

# Display the dropdown and plot the scatterplot
widgets.interactive(plot_scatterplot, genre=genre_dropdown)

# !jupyter nbconvert --to html Investigate_a_Dataset.ipynb

# Embed widgets
# !jupyter nbconvert --to html --execute Investigate_a_Dataset.ipynb

# Save requirements.txt file
# !pigar generate

	id	imdb_id	popularity	budget	revenue	original_title	cast	homepage	director	tagline	keywords	overview	runtime	genres	production_companies	release_date	vote_count	vote_average	release_year	budget_adj	revenue_adj
0	135397	tt0369610	32.986	150000000	1513528810	Jurassic World	Chris Pratt\|Bryce Dallas Howard\|Irrfan Khan\|Vi...	http://www.jurassicworld.com/	Colin Trevorrow	The park is open.	monster\|dna\|tyrannosaurus rex\|velociraptor\|island	Twenty-two years after the events of Jurassic ...	124	Action\|Adventure\|Science Fiction\|Thriller	Universal Studios\|Amblin Entertainment\|Legenda...	6/9/15	5562	6.500	2015	137999939.280	1392445892.524
1	76341	tt1392190	28.420	150000000	378436354	Mad Max: Fury Road	Tom Hardy\|Charlize Theron\|Hugh Keays-Byrne\|Nic...	http://www.madmaxmovie.com/	George Miller	What a Lovely Day.	future\|chase\|post-apocalyptic\|dystopia\|australia	An apocalyptic story set in the furthest reach...	120	Action\|Adventure\|Science Fiction\|Thriller	Village Roadshow Pictures\|Kennedy Miller Produ...	5/13/15	6185	7.100	2015	137999939.280	348161292.489
2	262500	tt2908446	13.113	110000000	295238201	Insurgent	Shailene Woodley\|Theo James\|Kate Winslet\|Ansel...	http://www.thedivergentseries.movie/#insurgent	Robert Schwentke	One Choice Can Destroy You	based on novel\|revolution\|dystopia\|sequel\|dyst...	Beatrice Prior must confront her inner demons ...	119	Adventure\|Science Fiction\|Thriller	Summit Entertainment\|Mandeville Films\|Red Wago...	3/18/15	2480	6.300	2015	101199955.472	271619025.408
3	140607	tt2488496	11.173	200000000	2068178225	Star Wars: The Force Awakens	Harrison Ford\|Mark Hamill\|Carrie Fisher\|Adam D...	http://www.starwars.com/films/star-wars-episod...	J.J. Abrams	Every generation has a story.	android\|spaceship\|jedi\|space opera\|3d	Thirty years after defeating the Galactic Empi...	136	Action\|Adventure\|Science Fiction\|Fantasy	Lucasfilm\|Truenorth Productions\|Bad Robot	12/15/15	5292	7.500	2015	183999919.040	1902723129.802
4	168259	tt2820852	9.335	190000000	1506249360	Furious 7	Vin Diesel\|Paul Walker\|Jason Statham\|Michelle ...	http://www.furious7.com/	James Wan	Vengeance Hits Home	car race\|speed\|revenge\|suspense\|car	Deckard Shaw seeks revenge against Dominic Tor...	137	Action\|Crime\|Thriller	Universal Pictures\|Original Film\|Media Rights ...	4/1/15	2947	7.300	2015	174799923.088	1385748801.471

	count	mean	std	min	25%	50%	75%	max
id	10866.000	66064.177	92130.137	5.000	10596.250	20669.000	75610.000	417859.000
popularity	10866.000	0.646	1.000	0.000	0.208	0.384	0.714	32.986
budget	10866.000	14625701.094	30913213.831	0.000	0.000	0.000	15000000.000	425000000.000
revenue	10866.000	39823319.793	117003486.582	0.000	0.000	0.000	24000000.000	2781505847.000
runtime	10866.000	102.071	31.381	0.000	90.000	99.000	111.000	900.000
vote_count	10866.000	217.390	575.619	10.000	17.000	38.000	145.750	9767.000
vote_average	10866.000	5.975	0.935	1.500	5.400	6.000	6.600	9.200
release_year	10866.000	2001.323	12.813	1960.000	1995.000	2006.000	2011.000	2015.000
budget_adj	10866.000	17551039.823	34306155.723	0.000	0.000	0.000	20853251.084	425000000.000
revenue_adj	10866.000	51364363.253	144632485.040	0.000	0.000	0.000	33697095.717	2827123750.412

	count	mean	std	min	25%	50%	75%	max
id	10865.000	66066.374	92134.092	5.000	10596.000	20662.000	75612.000	417859.000
popularity	10865.000	0.646	1.000	0.000	0.208	0.384	0.714	32.986
runtime	10865.000	102.072	31.383	0.000	90.000	99.000	111.000	900.000
vote_count	10865.000	217.400	575.645	10.000	17.000	38.000	146.000	9767.000
vote_average	10865.000	5.975	0.935	1.500	5.400	6.000	6.600	9.200
release_year	10865.000	2001.322	12.813	1960.000	1995.000	2006.000	2011.000	2015.000
budget_adj	10865.000	17549894.037	34307526.658	0.000	0.000	0.000	20853251.084	425000000.000
revenue_adj	10865.000	51369001.759	144638333.129	0.000	0.000	0.000	33701729.009	2827123750.412

	count	unique	top	freq
original_title	10865	10571	Hamlet	4
cast	10789	10719	Louis C.K.	6
director	10821	5067	Woody Allen	45
genres	10842	2039	Comedy	712
production_companies	9835	7445	Paramount Pictures	156

id	popularity	original_title	cast	director	runtime	genres	production_companies	release_date	vote_count	vote_average	release_year	budget_adj	revenue_adj
135397	32.986	Jurassic World	Chris Pratt	Colin Trevorrow	124.000	Action\|Adventure\|Science Fiction\|Thriller	Universal Studios\|Amblin Entertainment\|Legenda...	2015-06-09	5562	6.500	2015	137999939.280	1392445892.524
135397	32.986	Jurassic World	Bryce Dallas Howard	Colin Trevorrow	124.000	Action\|Adventure\|Science Fiction\|Thriller	Universal Studios\|Amblin Entertainment\|Legenda...	2015-06-09	5562	6.500	2015	137999939.280	1392445892.524
135397	32.986	Jurassic World	Irrfan Khan	Colin Trevorrow	124.000	Action\|Adventure\|Science Fiction\|Thriller	Universal Studios\|Amblin Entertainment\|Legenda...	2015-06-09	5562	6.500	2015	137999939.280	1392445892.524

Project: Investigate a Dataset - TMDB 5000 Movie Dataset¶

Selman Karaosmanoglu¶

Table of Contents¶

Introduction¶

Dataset Description¶

Question(s) for Analysis¶

Data Wrangling¶

Data Cleaning¶

User-defined functions¶

1. Drop unused columns¶

2. Remove duplicated columns¶

3. Check missing values¶

4. Convert to datetime format¶

Exploratory Data Analysis¶

User-defined functions¶

Research Question 1 - What are the top 5 cast members by frequency in the dataset?¶

Research Question 2 - What are the most and least frequent genres of movies in the dataset?¶

Research Question 3 - Which genres are most popular from year to year?¶

Research Question 4 - How many movies were released in each year in the dataset?¶

Research Question 5 - What is the median popularity of movies by genres?¶

Research Question 6 - Is there a statistically significant difference between median of the movie's popularity across genres?¶

Research Question 7 - Does the runtime of a movie have a different impact on its popularity depending on its genre?¶

Conclusions¶

Limitations¶

	id
genres
Action	2384
Adventure	1471
Animation	699
Comedy	3793
Crime	1354
Documentary	520
Drama	4760
Family	1231
Fantasy	916
Foreign	188
History	334
Horror	1637
Music	408
Mystery	810
Romance	1712
Science Fiction	1229
TV Movie	167
Thriller	2907
War	270
Western	165

	release_year	genres	popularity
0	1960	Action	0.591
1	1960	Adventure	0.701
2	1960	Comedy	0.396
3	1960	Crime	0.346
4	1960	Drama	0.566
5	1960	Family	0.278
6	1960	Fantasy	0.428
7	1960	Foreign	0.195
8	1960	History	0.412
9	1960	Horror	0.591
10	1960	Music	0.424
11	1960	Romance	0.541
12	1960	Science Fiction	0.328
13	1960	Thriller	0.812
14	1960	War	0.226
15	1960	Western	0.568
16	1961	Action	0.366
17	1961	Adventure	0.767
18	1961	Animation	2.632
19	1961	Comedy	0.680

	popularity
	mean	median
genres
Action	0.926	0.470
Adventure	1.154	0.561
Animation	0.852	0.501
Comedy	0.593	0.389
Crime	0.745	0.476
Documentary	0.181	0.142
Drama	0.591	0.389
Family	0.787	0.470
Fantasy	0.993	0.530
Foreign	0.191	0.165
History	0.576	0.413
Horror	0.465	0.322
Music	0.487	0.322
Mystery	0.690	0.428
Romance	0.592	0.404
Science Fiction	1.002	0.440
TV Movie	0.271	0.231
Thriller	0.742	0.444
War	0.728	0.432
Western	0.591	0.328

	release_year	genres	popularity
13	1960	Thriller	0.812
18	1961	Animation	2.632
34	1962	Adventure	0.943
52	1963	Animation	2.180
83	1964	War	0.931
95	1965	Music	0.969
105	1966	Animation	0.586
123	1967	Animation	1.349
149	1968	Mystery	1.519
159	1969	Crime	0.948
175	1970	Animation	1.128
199	1971	Family	1.531
215	1972	Crime	1.073
229	1973	Animation	0.957
257	1974	Mystery	0.702
265	1975	Adventure	0.880
285	1976	Crime	0.707
301	1977	Action	1.419
331	1978	Music	0.680
339	1979	Action	1.410
368	1980	Science Fiction	0.897
374	1981	Adventure	0.876
409	1982	War	1.143
412	1983	Adventure	0.901
437	1984	Family	0.824
455	1985	Family	0.924
467	1986	Adventure	0.799
491	1987	History	0.816
501	1988	Action	0.599
522	1989	Animation	1.178
540	1990	Adventure	0.802
561	1991	Animation	1.665
580	1992	Animation	1.287
605	1993	Fantasy	0.919
620	1994	Crime	1.298
638	1995	Animation	1.468
659	1996	Crime	0.977
689	1997	Science Fiction	1.140
711	1998	War	1.247
714	1999	Adventure	1.012